RSC Advances

View Article Online PAPER View Journal | View Issue

Rational re-design of Lactobacillus reuteri 121 inulosucrase for chain length control† Cite this: RSC Adv.,2019,9, 14957 Thanapon Charoenwongpaiboon,a Methus Klaewkla,ab Surasak Chunsrivirot,ab Karan Wangpaiboon,a Rath Pichyangkura,a Robert A. Field c and Manchumas Hengsakul Prousoontorn *a

Fructooligosaccharides (FOSs) are well-known prebiotics that are widely used in the food, beverage and pharmaceutical industries. Inulosucrase (E.C. 2.4.1.9) can potentially be used to synthesise FOSs from sucrose. In this study, inulosucrase from Lactobacillus reuteri 121 was engineered by site-directed mutagenesis to change the FOS chain length. Three variants (R483F, R483Y and R483W) were designed, and their binding free energies with 1,1,1-kestopentaose (GF4) were calculated with the Rosetta software. R483F and R483Y were predicted to bind with GF4 better than the wild type, suggesting that these engineered should be able to effectively extend GF4 by one residue and produce a greater quantity of GF5 than the wild type. MALDI-TOF MS analysis showed that R483F, R483Y and R483W variants Creative Commons Attribution-NonCommercial 3.0 Unported Licence. could synthesise shorter chain FOSs with a degree of polymerization (DP) up to 11, 10, and 10, respectively, while wild type produced longer FOSs and in polymeric form. Although the decrease in catalytic activity and the increase of hydrolysis/transglycosylation activity ratio was observed, the variants could effectively Received 20th March 2019 synthesise FOSs with the yield up to 73% of . Quantitative analysis demonstrated that these Accepted 7th May 2019 variants produced a larger quantity of GF5 than wild type, which was in good agreement with the predicted DOI: 10.1039/c9ra02137j binding free energy results. Our findings demonstrate the success of using aromatic amino acid residues, at rsc.li/rsc-advances position D418, to block the oligosaccharide binding track of inulosucrase in controlling product chain length. This article is licensed under a Introduction oligosaccharide ratio, synthesised by Bacillus megaterium levansucrase could be controlled by site-directed mutagen- Fructooligosaccharides (FOSs) are well-known prebiotics that esis.10,11 Furthermore, rationally designed variants of Bacillus Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM. are widely used in food, beverage and pharmaceutical applica- licheniformis levansucrase can increase the yield of oligosac- tions. FOSs are generally synthesised from sucrose using b- charides (DP4-25) up to 60% of the total products, and also alter fructofuranosidase (E.C. 3.2.1.26) derived from fungi, such as the ratio of isomeric trisaccharide products.12 Aspergillus niger,1,2 Aspergillus japonicas,3,4 Aspergillus aculeatus5,6 Previously, we reported the amino acid residues of Lactoba- and Aspergillus oryzae.7 Bacterial enzymes, such as levansucrase cillus reuteri 121 inulosucrase (LrInu) that play an essential role (E.C. 2.4.1.10) and inulosucrase (E.C. 2.4.1.9), have also been in FOS chain length determination.13 Some variants of this used to synthesise FOSs starting from sucrose producing b-2,6 , such as D479A, S482A, and R483A, can produce a broad linked levan and b-2,1 linked inulin, respectively.8,9 The bio- range of oligosaccharides with a small amount of accumulating logical activity of FOSs is largely dependent on their degree of polymer, while N543A mainly produces the short chain FOSs polymerization (DP). Many studies have attempted to engineer (DP3-6, with traces of DP $ 7). Since FOSs show optimal these enzymes to increase the yield of FOSs at the expense of prebiotic activity in the DP 2–8 range,14 it is interesting to use macromolecular polymer formation. In the case of levansu- the N543A variant of LrInu as a biocatalyst for the synthesis of crase, for example, the size of levan, as well as the polymer/ such oligosaccharides. However, this variant produced low amounts of transglycosylated product, but liberates high amounts of fructose from sucrose hydrolysis. Therefore, in this aDepartment of Biochemistry, Faculty of Science, Chulalongkorn University, study, we engineered inulosucrase to change its product chain Pathumwan, Bangkok, 10330, Thailand. E-mail: [email protected]; length using computer-aided rational design. The position of [email protected]; [email protected] bStructural and Computational Biology Research Unit, Department of Biochemistry, LrInu selected for mutagenesis was based on the oligosaccha- 13 Faculty of Science, Chulalongkorn University, Pathumwan, Bangkok, 10330, Thailand ride binding track of LrInu reported in previous study. The cDepartment of Biological Chemistry, John Innes Centre, Norwich Research Park, Rosetta program was employed using the homology model of Norwich, NR4 7UH, UK LrInu, built from crystal structure of Lactobacillus johnsonii † Electronic supplementary information (ESI) available. See DOI: inulosucrase (PDB ID: 2YFS),15 to assess variants of LrInu to F, Y 10.1039/c9ra02137j

This journal is © The Royal Society of Chemistry 2019 RSC Adv.,2019,9, 14957–14965 | 14957 View Article Online RSC Advances Paper

or W, to predict the substrate binding conformations and to noticed that the size of oligosaccharides synthesised by the D compute the binding free energies ( Gbinding) of the substrate/ variants did not correlate well with the distance between the enzyme complexes. The enzyme activities, biochemical prop- mutated sites and catalytic sites. In the case of LrInu, for erties and kinetic parameters of LrInu variants were determined example, the N543A variant mainly produced FOSs with DP3-6, and compared to that of the wild type. Finally, the FOSs while the R483A variant, whose R483 was located close to N543, produced by variant enzymes were analysed by TLC, HPLC and produced the FOSs up to DP12. Although these two residues MALDI-TOF MS. As predicted by Rosetta program, this study were close, the sizes of their FOSs products were dramatically demonstrated the effectiveness of using aromatic amino acids different. This nding suggested that there was more than one to block the oligosaccharide binding track and to control the amino acid residue located in the same that played size of oligosaccharides synthesised by inulosucrase. an important role in substrate binding. Although one of the binding residues was mutated, other residues still could hold Results and discussion the substrate, allowing transglycosylation to occur. Previous study found that blocking the oligosaccharide Rational protein design binding track of levansucrase with aromatic residues (F, Y and Many studies have shown that specicity of carbohydrate- W) was an effective strategy to block elongation of poly- modifying enzymes can be improved by enzyme engineering. saccharide and increase the yields of oligosaccharides.18 The size of glycan produced may be regulated by using two Therefore, we employed this approach in redesigning LrInu so different strategies: removing the interactions between that it could produce high yields of short to medium chain substrates and the enzymes' binding sites or blocking the length oligosaccharides. Because R483 of LrInu is located next substrate binding track with some amino acids, for instance to N543, we hypothesized that blocking at this position might aromatic amino acids (Fig. 1). Previously, many researches increase the yield of short FOSs like that obtained from the demonstrated that a single mutation at some specic amino N543A variant. To test this hypothesis, R483 residue of the

Creative Commons Attribution-NonCommercial 3.0 Unported Licence. acid residues of LrInu affected the size of inulin oligosaccharide catalytically competent binding conformation (GF4/Fru-WT) that could be produced.13,16,17 The yield of short-chain oligo- was changed in silico to A, F, Y and W to create GF4/Fru- saccharides was signicantly improved when some amino acid R483A, GF4/Fru-R483F, GF4/Fru-R483Y and GF4/Fru-R483W residues of LrInu were mutated to be alanine, which may reduce complexes, respectively, using the Rosetta program. This cata- the interactions between the enzyme and substrates with lytically competent binding conformation (GF4/Fru-WT) was medium or long chain lengths. This nding has also been re- dened as a binding conformation of 1,1,1-kestopentaose (GF4) ported on other fructosyltransferases, such as Bacillus mega- in the of Lactobacillus reuteri 121 inulosucrase con- terium levansucrase11 and Bacillus licheniformis levansucrase.12 taining fru-D272 (also see Experimental section). Since our However, those variants usually produced either very short previous study found that the R483A variant synthesised more This article is licensed under a range of FOSs (DP 3–6) or long range of FOSs (DP $ 12). To GF5 (DP6) than the wild type, the GF4/Fru-R483A complex was control the size of medium range of FOSs at DP # 10 is still used as positive control in this study. Fiy independent runs of challenging for the eld of enzyme engineering and food the FastDesign protocol were employed to resolve unfavorable

Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM. industry. Therefore, in this study, inulosucrase was redesigned interactions and nd low energy binding conformations of all D by computationally assisted method in order to increase the complexes (Fig. 2). Gbinding of each binding conformation was specicity of FOS produced in medium range. computed, and the average values are shown in Table 1. We From previous study, D479, S482, R483 and N543 residues hypothesized that there might be a possible correlation between were predicted to be the carbohydrate binding residues of the quantity of GF5 (DP6) product and the recognition of GF4 Lactobacillus reuteri 121 inulosucrase.13 Nevertheless, we (DP5). If an enzyme binds well with GF4, it should be able to effectively extend GF4 by one fructosyl residue to form GF5 via transfructosylation. As shown in Table 1, the Fru-R483A was predicted to bind GF4 better than the Fru-WT, suggesting a possible correlation between the quantity of GF5 products and

DGbinding of GF4 in the active site of Fru-R483A LrInu. Furthermore, our results show that the average values of

DGbinding of other variants are better than or about the same as that of the wild type, suggesting that they may be able to syn- thesise more quantity of GF5 products than the wild type as well. Therefore, the R483F, R483Y and R483W variants were selected for further kinetic study and product characterisations. When this study was conducted, the crystal structure of Lactobacillus reuteri 121 inulosucrase was not available. There- fore, its homology model was created in the previous study13 based on the crystal structure of Lactobacillus johnsonii inulo- 15 Fig. 1 Schematic display of enzyme engineering for modulation of the sucrase (PDB ID: 2YFS) and also used in this study. The size of oligosaccharide produced by . sequence identity of these two enzymes are reasonable with the

14958 | RSC Adv.,2019,9, 14957–14965 This journal is © The Royal Society of Chemistry 2019 View Article Online Paper RSC Advances Creative Commons Attribution-NonCommercial 3.0 Unported Licence. This article is licensed under a Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM.

Fig. 2 The predicted catalytically competent binding conformations of (A) wild type (B) R483A (C) R483F (D) R483Y and (E) R483W variants by Rosetta. The coordinates of all models are in ESI.†

value of 74.17%. Moreover, the quality of this homology model quality because the majority of amino acid residues are in was evaluated by Ramachandran plots in the previous study. favored region (93.7%) and allowed region (5.6%). Additionally, The results show that this homology model has promising the catalytic residues (D272, D424, and E523) of the homology model of inulosucrase from Lactobacillus reuteri 121 are in Table 1 The average values of predicted binding free energy similar positions to those of inulosucrase from Lactobacillus DG ( binding) of GF4 in the active sites of the wild-type and variant LrInu johnsonii, and they should be in appropriate positions for the containing a fructosyl-D272 intermediate (Fru-WT, Fru-R483A, Fru- catalysis of transfructosylation. Moreover, other studies also R483F, Fru-R483Y and Fru-R483W) calculated by Rosettaa employed homology models in creating structures of variant D 19–21 Gbinding (REU) proteins using the Rosetta program. In any case, if a crystal structure of Lactobacillus reuteri 121 inulosucrase is available in System Average s.e.m. the future and used as an input for the Rosetta program, more GF4/Fru-WT 6.3 0.1 accurate results might be obtained. GF4/Fru-R483A 7.1 0.1 GF4/Fru-R483F 7.0 0.1 GF4/Fru-R483Y 7.0 0.2 Activity and kinetic study of inulosucrase variants GF4/Fru-R483W 6.3 0.2 The designed LrInu variants above were expressed and puried a REU ¼ Rosetta energy unit. by the method described previously.13 Activity assays revealed

This journal is © The Royal Society of Chemistry 2019 RSC Adv.,2019,9, 14957–14965 | 14959 View Article Online RSC Advances Paper

that the engineered proteins had lower hydrolysis and trans- glycosylation activity than that of wild type (Table 2). In addi- tion, the hydrolysis/transglycosylation ratio was changed. The percentage of hydrolysis activity of the variant proteins was signicantly increased, up to 75% of total activity (Fig. 3). In comparison to the previous studies, the variants of fruc- tosyltransferase that produced the higher yield of oligosaccha- rides usually have a higher hydrolysis activity, such as, Y246A, N251A, K372A, R369A, R369S and R369K variants of levansu- crase from Bacillus licheniformis 8-37-0-1,12 N252A and K373R variant of levansucrase from Bacillus megaterium,11 and N543S variant of inulosucrase from Lactobacillus reuteri 121.17 The kinetic parameters of the wild-type and variant LrInu were determined based on activity versus sucrose concentration curves, tted with Hill and Michaelis–Menten equations.13,22 In previous study, we reported the kinetic parameters of wild-type LrInu. The total activity (VG) of wild-type LrInu was tted with – Hill equation, while the transglycosylation activity (VG F) was Fig. 3 Analysis of the H/T activity. The activity of wild-type and variant m 1 tted with a Michaelis–Menten equation, and hydrolysis activity inulosucrase was measured using 3.6 gmL enzyme and 250 mM sucrose, in 50 mM acetate buffer (pH 5.5) and 1 mM CaCl at 50 C. F  13 2 (V ) was tted with a substrate inhibition model. In this study, The blue columns represented the hydrolysis activity (H), while the red the kinetic behaviour of R483F, R483Y and R483W was also columns represented the transfructosylation activity (T). determined. The results demonstrated that the total activity G G–F F

Creative Commons Attribution-NonCommercial 3.0 Unported Licence. (V ), transglycosylation activity (V ) and hydrolysis activity (V ) optimum pH of all of the variant enzymes was not signicantly  – of all variants were best tted with Hill equation, Michaelis changed, while the optimum temperature of R483F, R483Y and Menten equation, and substrate inhibition model, respectively, R483W was signicantly shied from 50–60 Cto40–50 C indicating that the variant LrInu exhibited the same kinetic (Fig. 4). The change in optimal temperature of inulosucrase behaviour as the wild type. As shown in Table 2, the turnover might be resulted from the change of or G F G–F rate of all activities (kcat, kcat and kcat ) of R483F, R483Y and stability aer mutation. However, in practise, we usually syn-  R483W inulosucrases were signi cantly decreased compared to thesised the FOSs at sub-optimal temperature because of that of wild type, while the K50/m values of these inulosucrase a higher transglycosylation activity and stability. At high ffi variants were increased. Thus, the reduction of catalytic e - temperature, inulosucrase usually synthesised high amount of This article is licensed under a ciency (kcat/K50/m) of R483F, R483Y and R483W variants were fructose due to the increase in hydrolysis activity. observed. The reduction of catalytic activity of variants might be result from the conformational change of the enzyme cavity, 0

Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM.  which possibly in uence the proximity between C2 atom of Synthesis of FOSs by engineered inulosucrase fructosyl intermediate and O10 atom of acceptors. To synthesise FOSs, the engineered inulosucrase was incubated The effects of pH and temperature on enzyme activities were with 0.5 M sucrose in 50 mM acetate buffer (pH 5.5) at 30 C for also investigated in the pH and temperature ranges of 3.6–8.0 24 h. The reactions were performed at this sub-optimum and 10–70 C, respectively. The results showed that the temperature since it provides the enzyme higher

Table 2 Specific activity and kinetic parameters of WT and variant inulosucrase

WTa R483F R483Y R483W

Specic activity Total U mg 1 530 14 123 3 157 6 125 7 Hydrolysis U mg 1 282 25 90.2 2.4 117 5 88.4 2.4 Transglycosylation U mg 1 248 19 32.7 3.5 40.0 3.9 36.3 5.5 G 1 Kinetic parameter kcat s 2060 530 876 460 691 389 608 373 G K50/m mM nd nd nd nd G G 1 1 1 kcat (K50/m) mM s nd nd nd nd Hill factor 0.44 0.04 0.36 0.03 0.41 0.07 0.41 0.07 F 1 kcat s 523 62 120 3 153 5 119 3 F K50/m mM 25.3 7.3 26.5 3.0 24.7 3.9 27.6 2.8 F F 1 1 1 kcat (K50/m) mM s 20.7 4.52 6.18 4.31 G–F 1 kcat s 1830 300 154 35 269 76 176 35 G–F K50/m mM 1200 280 638 250 1110 470 733 241 G–F G–F 1 1 1 kcat (K50/m) mM s 1.52 0.241 0.242 0.240 a Charoenwongpaiboon et al. (2019).13 nd ¼ the result cannot be determined.

14960 | RSC Adv.,2019,9, 14957–14965 This journal is © The Royal Society of Chemistry 2019 View Article Online Paper RSC Advances

affected the yield of total FOSs. Furthermore, although the catalytic activity of variant enzymes decreased, it can be compensated by adding more biocatalyst to reach sufficient enzyme activity. Despite the fact that this strategy might increase the cost of FOS synthesis, these variant LrInu are still useful since they produce higher valued product (bioactive FOSs). In addition, the FOS products are easier to purify, hence, the cost of purication can be reduced. The amounts of some identied FOSs of variant and wild- type LrInu were also determined. It was found that R483F, R483Y and R483W produced a higher amount of some FOS species when compared to that of wild type (Fig. 6B). R483F produced a higher amount of DP5-8, whereas R483Y syn- thesised higher yield of DP4-8. Moreover, R483W also increased the yield of FOSs with DP3-7 when compared to that of the wild type. These ndings supported our conclusions from compu- tational studies which suggested that substitution by aromatic side chain would alter the product chain length specicity of LrInu. D According to the predicted Gbinding in Table 1, Fru-R483F and Fru-R483Y have better binding affinity to substrate GF4 (DP5) than the wild type, which may promote trans-

Creative Commons Attribution-NonCommercial 3.0 Unported Licence. fructosylation between the fru-D272 intermediate and acceptor GF4 (DP5). This would result in the accumulation of product GF5 (DP6) and support the synthesis of longer oligosaccharides. D However, although the average value of Gbinding of the R483W variant is about the same as that of the wild type, its ability to produce DP6 was unexpectedly higher than that of the wild type. This might be the limitation of the computational protein Fig. 4 (A) Effect of pH on initial velocity of wild-type and variant design since no protein design soware guarantees 100% inulosucrase. The reactions were performed using 3.6 mgmL1 successful results. Therefore, it should be conrmed by the in

This article is licensed under a ff – – enzymes and 250 mM sucrose in acetate bu er (pH 3.5 6.0) and bis vitro studies. However, in the author's opinion, this method- Tris buffer (pH 6.0–8.0) at 50 C. (B) Effect of temperature on initial ology is still useful for protein engineering and might have velocity of wild-type and variant inulosucrase. The reactions were  performed using 3.6 mgmL 1 enzymes and 250 mM sucrose in acetate a potential application for improving product speci city of Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM. buffer (pH 5.5) at the temperature range of 10–70 C. other enzymes.

Experimental transglycosylation activity and stability.13 TLC result demon- strated that the product patterns of all variant enzymes were Model construction and binding free energy calculation different (Fig. 5A). The variant enzymes synthesised a wide The structure of the catalytically competent binding confor- range of oligosaccharides, while the polymeric form was not mation of GF4 in the active site of Lactobacillus reuteri 121 observed. MALDI-TOF MS showed that wild type, R483A, R483F, inulosucrase containing a fructosyl-D272 intermediate (fru- R483Y and R483W produced the FOSs with the degree of poly- D272) was constructed from crystal structure of Lactobacillus merization (DP) up to 22 (m/z ¼ 3606.78), 12 (m/z ¼ 1985.48), 11 johnsonii inulosucrase (PDB ID: 2YFS with 74.17% identity),15 (m/z ¼ 1823.46), 10 (m/z ¼ 1661.47) and 10 (m/z ¼ 1661.47), according to method described in previous study,13 and was respectively (Fig. 5B). The size of FOSs decreased with the used as a template for mutations. The coordinates of this model increase of molecular weight of aromatic side chains, suggest- are in ESI.† This catalytically competent binding conformation ing that the bigger aromatic residue may block the enzyme's (GF4/Fru-WT) was dened as a binding conformation of GF4 in binding track better. the active site of Lactobacillus reuteri 121 inulosucrase con- Quantitative analysis of FOSs was performed by HPLC. The taining fru-D272, where the enzyme could potentially extend result showed that the total transglycosylation products (total GF4 by one fructosyl residue via transfructosylation. This FOSs) of variant LrInu were slightly decreased when compared binding conformation has the O1 atom of the non-reducing end to that of wild type (Fig. 6A). The transglycosylation products of of GF4 turning towards the C2 atom of the fructosyl residue of R483A, R483F, R483Y and R483W were approximately 73%, fru-D272, where the distance between these two atoms (O1–C2) 73%, 68% and 71% of total carbohydrate, respectively, while is reasonable.13 The FastDesign protocol23 of Rosetta 3.6 24 with that of wild type was 76%. This nding indicated that the the talaris2013 energy function25,26 was employed to predict and increase in hydrolysis activity of variant enzymes slightly optimize the binding conformations of GF4 in the active site of

This journal is © The Royal Society of Chemistry 2019 RSC Adv.,2019,9, 14957–14965 | 14961 View Article Online RSC Advances Paper Creative Commons Attribution-NonCommercial 3.0 Unported Licence. This article is licensed under a

Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM. Fig. 5 (A) TLC analysis and (B) MAL DI-TOF MS spectra of FOS products synthesised by the wild-type and variant LrInu. The reactions contained 0.5 M sucrose and 5 U mL1 of the enzymes in 50 mM acetate buffer (pH 5.5) at 30 C for 24 h.

Fig. 6 (A) The product specificity of wild-type and variant inulosucrase. (B) The amount of each FOS species produced by wild-type and variant inulosucrase. The reactions contained 0.5 M sucrose and 5 U mL1 of the enzymes in 50 mM acetate buffer (pH 5.5) at 30 C for 24 h.

14962 | RSC Adv.,2019,9, 14957–14965 This journal is © The Royal Society of Chemistry 2019 View Article Online Paper RSC Advances

the wild type, R483A, R483F, R483Y and R483W variants to an addition of 15 mL of 1 M NaOH. The total reducing sugar create GF4/Fru-WT, GF4/Fru-R483A, GF4/Fru-R483F, GF4/Fru- released by enzymes was determined by DNS assay,28 while the R483Y and GF4/Fru-R483W complexes, respectively. This total was determined by a glucose liquicolor kit protocol also employs the fast relax protocol to allow the (human™). The molar concentration of fructose produced from movement of sidechains and backbones to resolve energetically enzyme was calculated from the difference between the quality unfavorable features such as steric clashes, and subsequently of reducing sugar and glucose. One unit of total inulosucrase nd low energy conformations. Fiy independent runs of this activity was dened as the amount of enzyme required to release protocol19 were employed to produce 50 binding conformations 1 mmol of glucose per min, while hydrolysis activity was dened for the wild type, R483A, R483F, R483Y and R483W variants. as the amount of enzyme required to release 1 mmol of fructose Four separate sets of calculations were performed for the per min. Transglycosylation activity was calculated as the R483A, R483F, R483Y and R483W mutations. In each set, R483 difference between total activity and hydrolysis activity. was allowed to change to A, F, Y or W, respectively, and the The optimal pH of wild-type and variant LrInu was measured residues within 10 A˚ of residue 483 were repacked and mini- in the pH range of 3.6–8.0 at 50 C using the DNS assay. The mized. The free energy of each binding conformation (DG) was buffer systems used were 50 mM sodium acetate buffer (3.6–6.0)

calculated in Rosetta Energy Units (REU). DGbinding of each and 50 mM bis–Tris buffer (6.0–8.0). The optimal temperature binding conformation was calculated by subtracting the free of enzymes was determined in the temperature range of 30– energy of protein (DGprotein) and ligand (DGligand) from the free 70 C in 50 mM acetate buffer pH 5.5. D D ¼ D energy of the complex ( Gcomplex): Gbinding Gcomplex D D Gprotein Gligand. The details of binding free energy calcu- lation by Rosetta are in ESI.† The outlier data of each enzyme Enzyme kinetic were detected using SPSS soware and eliminated. The average Kinetic parameters of wild-type and variant inulosucrase were 13 values of DGbinding were then calculated. determined according to the method described previously. In m 1

Creative Commons Attribution-NonCommercial 3.0 Unported Licence. brief, the variant and wild-type inulosucrase (3.6 gmL ) were Construction, expression and purication of inulosucrase incubated with 500 mL of substrate solutions containing 0.5– 500 mM sucrose, 50 mM acetate buffer (pH 5.5) and 1 mM CaCl The construct InuD699His17 synthesised by Genscript, was used 2 at 50 C. The reactions were terminated by adding 15 mLof1M as a parental gene (wild-type) in this study. Site-directed muta- NaOH, and the amount of glucose and fructose released was genesis was performed by the PCR overlapping extension determined by the method described above. Activity versus method27 using PrimeStar™ DNA polymerase (Takara). The sucrose concentration curves were plotted and tted to either primers used for mutagenesis are described in the ESI section Hill or Michaelis–Menten equation using OriginPro soware in (Table S1†). The variant genes were ligated into pET-21b order to determine kinetic parameters of the enzymes (Fig. S1†). (Novagen™)atXhol and NdeI sites and were further trans- This article is licensed under a formed into E. coli strain Top10 (Invitrogen™) for cloning purposes. FOSs synthesis and characterisation The recombinant plasmids were transformed into E. coli 1 Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM. FOSs were synthesised using 5 U mL (glucose releasing BL21 (DE3). The E. coli carrying plasmid were cultured in LB activity) wild-type or variant inulosucrase, 500 mM sucrose, broth supplemented with 100 mgmL 1 ampicillin, 10 mM CaCl 2 50 mM acetate buffer (pH 5.5) and 1 mM CaCl . The reactions and 0.5% glucose, at 37 C, shaking at 250 rpm. Aer the cell 2 were incubated at 50 C for 24 h, then terminated by boiling for density reached (OD ) 0.4–0.6, IPTG was added to the nal 600 10 min. The resulting reaction mixtures were then analyzed by concentration of 0.1 mM. The cells were further cultured at TLC, HPLC and MALDI-TOF MS. 37 C, shaking at 200 rpm for 18–20 h. The cells were then The TLC system used in this study consisted of 1-buta- separated from media by centrifugation at 5000 g for 20 min nol : glacial acetic acid : water, 3 : 3 : 2 (v/v/v). The separation and lysed by ultra-sonication. The cell debris was removed from was performed by using TLC silica gel 60 F254 (Merck). The TLC crude enzyme by centrifugation at 12 000 g for 20 min. plates were dried and stained with a solution comprising 8 mL Crude enzymes were puried on a TOYOPEARL™ AF- of water, 10 mL of concentrated H SO , 27 mL of ethanol and Chelate-650M column pre-equilibrated with 25 mM potassium 2 4 0.1 g of orcinol. The TLC were visualized by heating. phosphate buffer (pH 7.4). The column was washed with the HPLC was performed on a Shimadzu™ (Prominence UFLC) same buffer containing 20 mM imidazole and 500 mM NaCl. instrument equipped with a refractive index detector. FOSs were Finally, the enzyme was eluted with 500 mM imidazole in the separated with an Asahipak NH2P-50 4E column (Shodex™) previous buffer. Protein concentrations were determined by using isocratic elution with 70% acetonitrile at a ow rate of 1 Bradford assay using a BSA as standard. mL min 1. As described in our previous study, the amount of identied FOSs was determined using the standard curve of Enzyme activity assay and biochemical characterisation glucose and fructose to determine monosaccharide, sucrose to Inulosucrase activity was measured by incubating the enzymes determine disaccharide, 1-kestose to determine trisaccharide, (3.6 mgmL 1) in 250 mM sucrose, containing 50 mM sodium and nystose to determine longer oligosaccharides.13,29

acetate buffer (pH 5.5) and 1 mM CaCl2, to reach the nal The mass of FOSs products was evaluated by MALDI-TOF volume of 500 mL, at 50 C. Then, the reaction was terminated by mass spectrometry (JEOL™ SpiralTOF MALDI Imaging-TOF/

This journal is © The Royal Society of Chemistry 2019 RSC Adv.,2019,9, 14957–14965 | 14963 View Article Online RSC Advances Paper

TOF Mass Spectrometer (JMS-S3000)). 2,5-Dihydroxybenzoic 6 L. Fernandez-Arrojo, B. Rodriguez-Colinas, P. Gutierrez- acid (DHB) was used as the matrix. Alonso, M. Fernandez-Lobato, M. Alcalde, A. O. Ballesteros and F. J. Plou, Process Biochem., 2013, 48, 677–682. 7 M. Kurakake, R. Masumoto, K. Maguma, A. Kamata, E. Saito, Conclusions N. Ukita and T. Komaki, J. Agric. Food Chem., 2010, 58, 488– 492. Employing computer-aided rational protein mutagenesis, we 8 L. K. Ozimek, S. Kralj, M. J. E. C. van der Maarel and successfully engineered LrInu variants (R483F, R483Y and L. Dijkhuizen, Microbiology, 2006, 152, 1187–1196. R483W variants) that can produce signicantly higher yields of 9 D. Ni, W. Zhang, C. Guang, T. Zhang and W. Mu, Crit. Rev. DP 4–8 oligosaccharides than the wild type. Our results indicate Food Sci. Nutr., 2018, 1–18, DOI: 10.1080/ that the yields of some FOS synthesised by inulosucrase variants 10408398.2018.1506421. correlate, to some extent, with the relative binding free energies 10 A. Homann, R. Biedendieck, S. Gotze,¨ D. Jahn and J. Seibel, predicted by the Rosetta program. Overall, this study showed Biochem. J., 2007, 407, 189–198. the success of using aromatic amino acids, as predicted by 11 C. P. Strube, A. Homann, M. Gamer, D. Jahn, J. Seibel and Rosetta program, to block the oligosaccharide binding track in D. W. Heinz, J. Biol. Chem., 2011, 286, 17593–17600. inulosucrase controlling the size of oligosaccharides syn- 12 C. He, Y. Yang, R. Zhao, J. Qu, L. Jin, L. Lu, L. Xu and M. Xiao, thesised and demonstrating the effectiveness of the designed Appl. Microbiol. Biotechnol., 2018, 102, 3217–3228. enzymes in producing high yields of useful FOSs. 13 T. Charoenwongpaiboon, T. Sitthiyotha, P. P. Na Ayutthaya, K. Wangpaiboon, S. Chunsrivirot, M. Hengsakul Conflicts of interest Prousoontorn and R. Pichyangkura, Carbohydr. Polym., 2019, 209, 111–121. There are no conicts to declare. 14 E. Biedrzycka and M. Bielecka, Trends Food Sci. Technol., –

Creative Commons Attribution-NonCommercial 3.0 Unported Licence. 2004, 15, 170 175. 15 T. Pijning, M. A. Anwar, M. Boger,¨ J. M. Dobruchowska, Acknowledgements H. Leemhuis, S. Kralj, L. Dijkhuizen and B. W. Dijkstra, J. Mol. Biol., 2011, 412,80–93. TC is thankful for scholarship supports from Science Achieve- 16 L. K. Ozimek, S. Kralj, T. Kaper, M. J. van der Maarel and ment Scholarship of Thailand (SAST). Supports from the Over- L. Dijkhuizen, FEBS J., 2006, 273, 4104–4113. seas Research Experience Scholarship for Graduate Student and 17 M. A. Anwar, H. Leemhuis, T. Pijning, S. Kralj, B. W. Dijkstra Faculty of Science Chulalongkorn University are acknowledged. and L. Dijkhuizen, FEBS J., 2012, 279, 3612–3621. MK gratefully acknowledged the Scholarship from the Graduate 18 P. Kanjanatanin, R. Pichyangkura, T. Sitthiyotha, School, Chulalongkorn University to commemorate the 72nd This article is licensed under a T. Charoenwongpaiboon, K. Wangpaiboon and anniversary of his Majesty King Bhumibol Adulyadej. SC was S. Chunsrivirot, 2019, under review. also partially supported by Structural and Computational 19 C. S. Poultney, G. L. Butterfoss, M. R. Gutwein, K. Drew, Biology Research Unit, Department of Biochemistry, Faculty of

Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM. D. Gresham, K. C. Gunsalus, D. E. Shasha and R. Bonneau, Science, Rachadaphiseksomphot Endowment Fund, Chula- PLoS One, 2011, 6, e23947. longkorn University, the Institute for the Promotion of Teaching 20 A. E. Miklos, C. Kluwe, S. D. Bryan, S. Pai, A. Sircar, Science and Technology (IPST) under the Research Fund for R. A. Hughes, M. Berrondo, J. Xu, V. Codrea, P. E. Buckley, DPST Graduate with First Placement [Grant no. 07/2557] and A. M. Calm, H. S. Welsh, C. R. Warner, M. A. Zacharko, Chulalongkorn University; Government Budget. Research at the J. P. Carney, J. J. Gray, G. Georgiou, B. Kuhlman and JIC is supported by the UK BBSRC Institute Strategic Program A. D. Ellington, Chem. Biol., 2012, 19, 449–455. on Molecules from Nature - Products and Pathways [BBS/E/J/ 21 Y. Xia, W. Cui, Z. Cheng, L. Peplowski, Z. Liu, M. Kobayashi 000PR9790] and the John Innes Foundation. and Z. Zhou, ChemCatChem, 2018, 10, 1370–1375. 22 S. A. F. T. van Hijum, M. J. E. C. van der Maarel and References L. Dijkhuizen, FEBS Lett., 2003, 534, 207–210. 23 G. Bhardwaj, V. K. Mulligan, C. D. Bahl, J. M. Gilmore, 1 C. Goosen, X.-L. Yuan, J. M. van Munster, A. F. J. Ram, P. J. Harvey, O. Cheneval, G. W. Buchko, M. J. E. C. van der Maarel and L. Dijkhuizen, Eukaryotic S. V. S. R. K. Pulavarti, Q. Kaas, A. Eletsky, P.-S. Huang, Cell, 2007, 6, 674–681. W. A. Johnsen, P. Greisen Jr, G. J. Rocklin, Y. Song, 2 K. Yanai, A. Nakane, A. Kawate and M. Hirayama, Biosci., T. W. Linsky, A. Watkins, S. A. Rettie, X. Xu, L. P. Carter, Biotechnol., Biochem., 2001, 65, 766–773. R. Bonneau, J. M. Olson, E. Coutsias, C. E. Correnti, 3 S. I. Mussatto, C. N. Aguilar, L. R. Rodrigues and T. Szyperski, D. J. Craik and D. Baker, Nature, 2016, 538, 329. J. A. Teixeira, J. Mol. Catal. B: Enzym., 2009, 59,76–81. 24 A. Leaver-Fay, M. Tyka, S. M. Lewis, O. F. Lange, 4 Y.-l. Yang, J.-h. Wang, D. Teng and F. Zhang, J. Agric. Food J. Thompson, R. Jacak, K. W. Kaufman, P. D. Renfrew, Chem., 2008, 56, 2805–2809. C. A. Smith, W. Sheffler, I. W. Davis, S. Cooper, A. Treuille, 5 A. S. G. Lorenzoni, L. F. Aydos, M. P. Klein, R. C. Rodrigues D. J. Mandell, F. Richter, Y.-E. A. Ban, S. J. Fleishman, and P. F. Hertz, Carbohydr. Polym., 2014, 103, 193–197. J. E. Corn, D. E. Kim, S. Lyskov, M. Berrondo, S. Mentzer,

14964 | RSC Adv.,2019,9, 14957–14965 This journal is © The Royal Society of Chemistry 2019 View Article Online Paper RSC Advances

Z. Popovi´c, J. J. Havranek, J. Karanicolas, R. Das, J. Meiler, 26 A. Bazzoli, S. P. Kelow and J. Karanicolas, PLoS One, 2015, 10, T. Kortemme, J. J. Gray, B. Kuhlman, D. Baker and e0140359. P. Bradley, in Methods in Enzymology, ed. M. L. Johnson 27 K. L. Heckman and L. R. Pease, Nat. Protoc., 2007, 2, 924. and L. Brand, Academic Press, 2011, vol. 487, pp. 545–574. 28 G. L. Miller, Anal. Chem., 1959, 31, 426–428. 25 M. J. O'Meara, A. Leaver-Fay, M. D. Tyka, A. Stein, 29 T. Charoenwongpaiboon, K. Wangpaiboon, R. Pichyangkura K. Houlihan, F. DiMaio, P. Bradley, T. Kortemme, and M. H. Prousoontorn, RSC Adv., 2018, 8, 17008–17016. D. Baker, J. Snoeyink and B. Kuhlman, J. Chem. Theory Comput., 2015, 11, 609–622. Creative Commons Attribution-NonCommercial 3.0 Unported Licence. This article is licensed under a Open Access Article. Published on 14 May 2019. Downloaded 9/27/2021 3:01:54 PM.

This journal is © The Royal Society of Chemistry 2019 RSC Adv.,2019,9, 14957–14965 | 14965